Maude-help mailing list

[Michael Katelman <katelman AT uiuc.edu>]

Pat,

Could you clarify exactly what the "expression problem" is? I am
having difficulty following the wikipedia entry.

Since you mentioned type classes (a la Haskell), perhaps it's relevant
to note that while ad-hoc overloading of functions is allowed in
Maude, the situation is somewhat different from Haskell's type
classes.  Haskell's type classes are used to support a sort of
restricted polymorphism where the type signature quantifies over a
subset of all types with the same overloaded operators, but the
"type-system" of Maude is not polymorphic; in the sense that the
signatures of functions cannot have universal quantifiers over
"types", like in Haskell.

At least that's my take on how the two are related.

-Mike

On Thu, Dec 31, 2009 at 5:44 AM, pbrowne 
<Patrick.Browne AT comp.dit.ie>
 wrote:
> hi,
> I have a question concerning Wadler's expression problem[1] for
> algebraic data types. In relation to Haskells type classes is easy to
> add a new operations on existing data types, it requires only the
> definition of a new function. All the old functions on those types
> continue to work unchanged. On the other hand, it is difficult to add
> new structure to an existing data type: it requires the addition of a
> new constructor for the existing data type.
>
> Questions:
> 1)Does expression problem exist in Maude?
> 2)If it does not exist in Maude, is there an illustrative example available?
>
>
> Pat
>
> [1]http://en.wikipedia.org/wiki/Expression_Problem
>
> _______________________________________________
> Maude-help mailing list
> Maude-help AT cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/maude-help
>